Zobrazeno 1 - 10
of 168
pro vyhledávání: '"Cisse, Moustapha"'
Variance reduction (VR) techniques have contributed significantly to accelerating learning with massive datasets in the smooth and strongly convex setting (Schmidt et al., 2017; Johnson & Zhang, 2013; Roux et al., 2012). However, such techniques have
Externí odkaz:
http://arxiv.org/abs/2111.11828
Autor:
Sirko, Wojciech, Kashubin, Sergii, Ritter, Marvin, Annkah, Abigail, Bouchareb, Yasser Salah Eddine, Dauphin, Yann, Keysers, Daniel, Neumann, Maxim, Cisse, Moustapha, Quinn, John
Identifying the locations and footprints of buildings is vital for many practical and scientific purposes. Such information can be particularly useful in developing regions where alternative data sources may be scarce. In this work, we describe a mod
Externí odkaz:
http://arxiv.org/abs/2107.12283
In algorithmically fair prediction problems, a standard goal is to ensure the equality of fairness metrics across multiple overlapping groups simultaneously. We reconsider this standard fair classification problem using a probabilistic population ana
Externí odkaz:
http://arxiv.org/abs/2006.13485
Mixup is a data augmentation technique that creates new examples as convex combinations of training points and labels. This simple technique has empirically shown to improve the accuracy of many state-of-the-art models in different settings and appli
Externí odkaz:
http://arxiv.org/abs/2006.06049
Deep Neural Networks have recently gained lots of success after enabling several breakthroughs in notoriously challenging problems. Training these networks is computationally expensive and requires vast amounts of training data. Selling such pre-trai
Externí odkaz:
http://arxiv.org/abs/1802.04633
Automatic speaker verification systems are increasingly used as the primary means to authenticate costumers. Recently, it has been proposed to train speaker verification systems using end-to-end deep neural models. In this paper, we show that such sy
Externí odkaz:
http://arxiv.org/abs/1801.03339
Autor:
Stock, Pierre, Cisse, Moustapha
ConvNets and Imagenet have driven the recent success of deep learning for image classification. However, the marked slowdown in performance improvement combined with the lack of robustness of neural networks to adversarial examples and their tendency
Externí odkaz:
http://arxiv.org/abs/1711.11443
Recently, continuous cache models were proposed as extensions to recurrent neural network language models, to adapt their predictions to local changes in the data distribution. These models only capture the local context, of up to a few thousands tok
Externí odkaz:
http://arxiv.org/abs/1711.02604
This paper investigates strategies that defend against adversarial-example attacks on image-classification systems by transforming the inputs before feeding them to the system. Specifically, we study applying image transformations such as bit-depth r
Externí odkaz:
http://arxiv.org/abs/1711.00117
Large deep neural networks are powerful, but exhibit undesirable behaviors such as memorization and sensitivity to adversarial examples. In this work, we propose mixup, a simple learning principle to alleviate these issues. In essence, mixup trains a
Externí odkaz:
http://arxiv.org/abs/1710.09412